Reaching for objects is a natural and common behavior. A great deal of behavioral and physiological effort has been devoted to understanding reaching behavior. This project is the continuation of a unique attempt to understand how humans are able to reach accurately despite the highly variable dynamics of their own movement patterns. Specifically, this project attempts to understand reaches involving simultaneous extension of the arm and torso rotation. The project was originally based on the observation that when we turn and reach for an object, the Coriolis forces generated between the torso and arm are orders of magnitude greater than the interaction forces generated between brachial segments during simple reaches not involving trunk rotation. If dynamic perturbations of this type and magnitude are applied to the arm by a robotic manipulandum, large movement errors result. Subjects can learn to move accurately again through repetition but the adaptation does not generalize far beyond the workspace region in which the mechanical perturbation was experienced. In the first five years of this project, we showed that 1) humans can create vastly different Coriolis forces during different reaches, 2) these force variations have no discernible effect on the accuracy of turn and reach movements, 3) this accuracy is achieved through both feedforward compensation for anticipated Coriolis forces and on-line compensations, 4) adaptive self re-calibration of feedforward mechanisms is possible, and 5) signals about foot-centric torso rotation are implicated in re-calibration. Our new aims include 1) understanding the control of posture and gaze during turn and reach movements and 2) discriminating feed forward and feedback mechanisms in reaching movements made during passive body rotation. The experimental findings and analyses will provide a comprehensive evaluation of the role of feedforward predictive mechanisms in the control of reaching movements involving body rotation.